Affective computing
| Hi-index | 0.00 |
Psychophysiological signals enable computer systems to monitor the emotional state of a user. Such a system could adapt its behavior to reduce stress, give assistance, or suggest well-being tips. All of this should lead to a technology that is more user-friendly and more accessible to older people. Measuring physiological signals in research labs has been done for many years. In such a controlled environment the quality of signals is very high because of the optimal placement of electrodes by research staff. Analysis techniques can therefore rely on high quality data. Measuring physiological signals in real-life settings without the assistance of well-trained staff, is much more challenging because of artifacts and signal distortions. In this paper we discuss the approach taken in the Aladin project to cope with the inferior and unreliable quality of physiological signal measurements. We discuss a sensor design intended for every-day use and present the variance of skin conductance we experienced within measurements, between different measurements of the same individual as well as between different persons. Finally, we suggest using trends instead of absolute values as a basis for physiology-enhanced human-computer interaction "in the wild".